Use of Colloidal Graphite Coating to Reduce Magnetic Resonance Imaging Artifacts Caused by Metallic Objects

Magnetic susceptibility mismatch, between human tissue and a foreign metallic object, is one of several factors responsible for image distortions in magnetic resonance imaging (MRI). Combining diamagnetic materials such as bismuth or carbon with paramagnetic materials such as nitinol or titanium can reduce the mismatch in bulk susceptibility of a foreign object and the surrounding tissue. Muller-Bierl et al. have succeeded in reducing MRI field distortion by coating titanium wire with bismuth. Wilson et al. used a pyrolytic graphite mouth shim to improve brain functional MRI performance. Conolly et al. have successfully used pyrolytic graphite in foam to reduce image artifacts at air-tissue interfaces. In this study, it was hypothesized that coating a metallic object with carbon particles suspended in a polymer can reduce the size of image artifacts. Four 6Al-4V titanium discs $(2.3mm×9.5mm∅)$ were encapsulated in an epoxy-graphite mixture. Mixtures of graphite and epoxy were poured around the titanium discs in molds and allowed to cure. A specimen of titanium was encapsulated in plain epoxy to serve as the control sample. Polycrystalline graphite was mixed at mass ratios of 1:2 and 1:1 to epoxy for two of the samples. Pyrolytic graphite flakes were mixed at a 1:2 mass ratio to epoxy. The sample discs were placed in an aqueous solution of copper sulfate and gadolinium contrast agent inside a wrist imaging coil at the isocenter of a 3 Tesla MRI machine; disc axes were perpendicular to the $B0$ direction. A T2-weighted gradient echo MRI image was taken in the coronal plane. Echo time, relaxation time, flip angle, and phase encode direction set to 71 ms, 3430 ms, 80 degrees, and right to left respectively. The control sample produced an arrowhead artifact sweeping in the same direction as the static magnetic field vector, $B0$⁠. The two samples containing powdered polycrystalline graphite produced arrowhead shaped artifacts. The direction of image distortion, however, was opposite from that of the control sample. The change in direction of the image artifact is attributed to the change in bulk magnetic susceptibility of the sample from paramagnetic behavior of titanium encapsulated in plain epoxy to a diamagnetic behavior from the added carbon powder. The titanium sample encapsulated in the pyrolytic graphite-epoxy mixture produced an artifact with irregular outline and no discernable directional bias relative to $B0$⁠. The hypothesized cause for this difference in artifact shape between the polycrystalline and pyrolytic graphite samples is an increase in air bubble entrapment due to the planar structure of the pyrolytic graphite flakes during the epoxy mixing process. Further study is underway to find a specific carbon-polymer mass ratio and coating thickness that will reduce MR image artifacts that would otherwise appear due to the presence of a metallic object in the MRI region of interest. This work is supported by MIMTeC, a National Science Foundation Industry University Collaborative Research Center and by NIH Grant P30NS057091.